东海初级生产力遥感反演及其时空演化机制

According to calculation results of ocean chlorophyll concentration based on SeaWiFS data by SeaBAM model and synchronous ship-measured data, this research set up an improved model for Case I and Case Ⅱ water bodies respectively. The monthly chlorophyll distribution in the East China Sea in 1998 was obtained from this improved model on calculation results of SeaBAM. The euphotic depth distribution in 1998 in the East China Sea is calculated by using remote sensing data of K490 from SeaWiFS according to the relation between the euphotic depth and the oceanic diffuse attenuation coefficient. With data of ocean chlorophyll concentration, euphotic depth, ocean surface photosynthetic available radiation (PAR), daily photoperiod and optimal rate of daily carbon fixation within a water column, the monthly and annual primary productivity spatio-temporal distributions in the East China Sea in 1998 were obtained based on VGPM model. Based on analysis of those distributions, the conclusion can be drawn that there is a clear bimodality character of primary productivity in the monthly distribution in the East China Sea. In detail, the monthly distribution of primary productivity stays the lowest level in winter and rises rapidly to the peak in spring. It gets down a little in summer, and gets up a little in autumn. The daily average of primary productivity in the whole East China Sea is 560.03 mg/m^2/d, which is far higher than the average of subtropical ocean areas. The annual average of primary productivity is 236.95 g/m^2/a. The research on the seasonal variety mechanism of primary productivity shows that several factors that affect the spatio-temporal distribution may include the chlorophyll concentration distribution, temperature condition, the Yangtze River diluted water variety, the euphotic depth, ocean current variety, etc. But the main influencing factors may be different in each local sea area.     

Estimation of ocean primary productivity and its spatio-temporal variation mechanism for East China Sea based on VGPM model

According to calculation results of ocean chlorophyll concentration based on SeaWiFS data by SeaBAM model and synchronous ship-measured data, this research set up an improved model for CaseⅠand CaseⅡwater bodies respectively. The monthly chlorophyll distribution in the East China Sea in 1998 was obtained from this improved model on calculation results of SeaBAM. The euphotic depth distribution in 1998 in the East China Sea is calculated by using remote sensing data of K490 from SeaWiFS according to the relation between the euphotic depth and the oceanic diffuse attenuation coefficient. With data of ocean chlorophyll concentration, euphotic depth, ocean surface photosynthetic available radiation (PAR), daily photoperiod and optimal rate of daily carbon fixation within a water column, the monthly and annual primary productivity spatio-temporal distributions in the East China Sea in 1998 were obtained based on VGPM model. Based on analysis of those distributions, the conclusion can be drawn that there is a clear bimodality character of primary productivity in the monthly distribution in the East China Sea. In detail, the monthly distribution of primary productivity stays the lowest level in winter and rises rapidly to the peak in spring. It gets down a little in summer, and gets up a little in autumn. The daily average of primary productivity in the whole East China Sea is 560.03 mg/m2/d, which is far higher than the average of subtropical ocean areas. The annual average of primary productivity is 236.95 g/m2/a. The research on the seasonal variety mechanism of primary productivity shows that several factors that affect the spatio-temporal distribution may include the chlorophyll concentration distribution, temperature condition, the Yangtze River diluted water variety, the euphotic depth, ocean current variety, etc. But the main influencing factors may be different in each local sea area.     

渤海海洋悬浮泥沙二元特征参数MODIS遥感反演模型研究

This paper brought out a new idea on the retrieval of suspended sediment concentration, which uses both the water-leaving radiance from remote sensing data and the grain size of the suspended sediment. A principal component model and a neural network model based on those two parameters were constructed. The analyzing results indicate that testing errors of the models using the two parameters are 0.256 and 0.244, while the errors using only water-leaving radiance are 0,384 and 0.390. The stability of the models with grain size parameter is also better than the one without grain size. This research proved that it is necessary to introduce the grain size parameter into suspended sediment concentration retrieval models in order to improve the retrieval precision of these models.     

DETECTING AND ADJUSTING FOR NON-LINEARITIES IN CALIBRATION OF NEAR-INFRARED DATA USING PRINCIPAL COMPONENTS

A new regression method for non-linear near-infrared spectroscopic data is proposed.The technique isbased on a model which is linear in the principal components and simple functions(squares and products)of them.Added variable plots are used to determine which squares and products to incorporate into themodel.The regression coefficients are estimated by a Stein estimate which shrinks towards the estimatedetermined by the first several principal components and the selected non-linear terms.The technique isnot computationally intensive and is appropriate for routine predictions of chemical concentrations.Themethod is tested on three data sets and in all cases gives more accurate predictions than does linearprincipal components regression.     

THE EFFECT OF MEAN CENTERING ON PREDICTION IN MULTIVARIATE CALIBRATION

Traditionally,one form of preprocessing in multivariate calibration methods such as principal componentregression and partial least squares is mean centering the independent variables(responses)and thedependent variables(concentrations).However,upon examination of the statistical issue of errorpropagation in multivariate calibration,it was found that mean centering is not advised for some datastructures.In this paper it is shown that for response data which(i)vary linearly with concentration,(ii)have no baseline(when there is a component with a non-zero response that does not change inconcentration)and(iii)have no closure in the concentrations(for each sample the concentrations of allcomponents add to a constant,e.g.100%)it is better not to mean center the calibration data.That is,the prediction errors as evaluated by a root mean square error statistic will be smaller for a model madewith the raw data than a model made with mean-centered data.With simulated data relativeimprovements ranging from 1% to 13% were observed depending on the amount of error in thecalibration concentrations and responses.     

A TEST OF THREE FITTING CRITERIA FOR MULTIRESPONSE NON-LINEAR MODELING

This work evaluates objective functions for multiresponse non-linear modeling using computersimulations.Tests are performed under a variety of signal-to-noise ratios and noise variance-covariancestructures.The standard error of prediction for the model parameters,computed from 50 trials,is usedfor performance comparisons.The full rank and rank-deficient problems are considered.For the fullrank problem one model was investigated,a first-order two-step consecutive reaction model,and twoobjective functions were considered,the total sum of squares and the determinant criterion.Nodistinction could be made between the two objective functions for this model.For the rank-deficient case two models were investigated,a first-order two-step consecutive reactionas in the full rank case,and a pH titration model described by the Henderson-Hasselbalch equation.Three objective functions were investigated for the rank-deficient case,the total sum of squares,aweighted total sum of squares and the determinant criterion.The total sum of squares was found toperform poorly under all conditions tested compared to the weighted total sum of squares and thedeterminant criterion.The determinant criterion was found to perform much better than the other twocriteria when the data have a combination of a low signal-to-noise ratio and high variance-covariancenoise structure.     

REPRESENTATION OF SPECTRA BY CONTINUOUS FUNCTIONS

A general framework for manipulating spectra as functions in traditional multivariate methods such asPCA and PLS is described.The functional representation is very convenient for compression,ensuringsmoothness and continuity.There are two fundamentally different types of representations:(a)byfunctions and(b)by function coefficients.The use of coefficients is the most practical way of analysis.     

Modeling dynamic assessment of ecosystem services based on remote sensing technology: A sampling of the Gansu grassland ecosystem

The ecosystem is important because it is the life sustaining system for human survival.Three ecosystem characteristics are:regional particularities,ecosystem complexity and conventional cultural particularities.This paper develops a remote sensing based dynamic model to assess grassland ecosystem service values involving multidisciplinary knowledge.The ecological value of grassland ecosystems is focused on using a remote sensing technique in the model,and setting up the framework for a dynamic assessing model.The grassland ecological services condition and value in 1985 is used as the benchmark.The dynamic model has two adjusting indicators:biomass and price index.The biomass is simulated using the CASA(Carnegie-Ames-Stanford Approach) model.The price index was obtained from statistics data published by the statistical bureau.Results show that the grassland ecosystem value in Gansu Province was 28.36 billion Chinese Yuan in 1985,140.37 billion in 1999 and 130.86 billion in 2002.     

CHEMICAL INFORMATION FROM ELECTROANALYTICAL DATA.PART 1—DETERMINATION OF SYSTEM PARAMETERS FOR QUASI-REVERSIBLE ELECTRON TRANSFER REACTIONS FROM CYCLIC VOLTAMMETRIC TEST DATA AND DATA FOR THE REDUCTION OF CERIUM(Ⅳ)BIS(OCTAETHYLPORPHYRINATE)

Values of the peak potential,the peak current function and the peak potential difference are extractedfrom simulated cyclic voltammograms of a quasi-reversible electron transfer reaction.Variations of thefeature values are described by empirical functions of the model parameters α(transfer coefficient)andψ'(dimensionless form of the heterogeneous electron transfer rate constant k_s)using multiple linearregression.These functions are used to determine the system parameters α and k_s from test data sets offeature values.The predictive power of the features is discussed.It is shown that several features can beused simultaneously in the analysis.Also,α,k_s,the formal potential E~0 and the diffusion coefficient Dcan be estimated simultaneously.Experimental data from the metal-centered one-electron reduction ofcerium(Ⅳ)bis(octaethylporphyrinate),Ce(OEP)_2,are analyzed using the technique,and values of thesystem parameters for this electrochemical reaction are determined.     

长江口水域悬沙浓度时空变化与泥沙再悬浮

A detailed analysis of suspended sediment concentration (SSC) variations over a year period is presented using the data from 8 stations in the Yangtze River estuary and its adjacent waters, together with a discussion of the hydrodynamic regimes of the estuary. Spatially, the SSC from Xuliujing downwards to Hangzhou Bay increases almost constantly, and the suspended sediment in the inner estuary shows higher concentration in summer than in winter, while in the outer estuary it shows higher concentration in winter than in summer, and the magnitude is greater in the outer estuary than in the inner estuary, greater in the Hangzhou Bay than in the Yangtze River estuary. The sediments discharged by the Yangtze River into the sea are resuspended by marine dynamics included tidal currents and wind waves. Temporally, the SSC shows a pronounced neap-spring tidal cycle and seasonal variations. Furthermore, through the analysis of dynamic mechanism, it is concluded that wave and tidal current are two predominant factors of sediment resuspension and control the distribution and changes of SSC, in which tidal currents control neap-spring tidal cycles, and wind waves control seasonal variations. The ratio between river discharge and marine dynamics controls soatial distribution of SSC.     

DATA ANALYSIS WITH MINIMAL ASSUMPTIONS

The computer has made it possible to scrutinize data rapidly by means of graphics. This should be doneprior to the application of any model to the data, since the model must be validated before using it asa means of analyzing the data. The procedure is illustrated in terms of two examples of real experimentaldata.     

Monthly calibration and optimization of ?ngstr?m-Prescott equation coefficients for comprehensive agricultural divisions in China

?ngstr?m-Prescott equation (AP) is the algorithm recommended by the Food and Agriculture Organization (FAO) of the United Nations for calculating the surface solar radiation (R_s) to support the estimation of crop evapotranspiration.Thus,the a_s and b_s coefficients in the AP are vital.This study aims to obtain coefficients a_s and b_s in the AP,which are optimized for China’s comprehensive agricultural divisions.The average monthly solar radiation and relative sunshine duration data at 121 stations from 1957–2016 were collected.Using data from 1957 to 2010,we calculated the monthly a_s and b_s coefficients for each subregion by least-squares regression.Then,taking the observation values of R_s from 2011 to 2016 as the true values,we estimated and compared the relative accuracy of R_s calculated using the regression values of coefficients a_s and b_s and that calculated with the FAO recommended coefficients.The monthly coefficients,a_s and b_s,of each subregion are significantly different,both temporally and spatially,from the FAO recommended coefficients.The relative error range (0–54%) of R_s calculated via the regression values of the a_s and b_s coefficients is better than the relative error range (0–77%) of R_s calculated using the FAO suggested coefficients.The station-mean relative error was reduced by 1%to 6%.However,the regression values of the a_s and b_s coefficients performed worse in certain months and agricultural subregions during verification.Therefore,we selected the a_s and b_s coefficients with the minimum R_(s )estimation error as the final coefficients and constructed a coefficient recommendation table for 36 agricultural production and management subregions in China.These coefficient recommendations enrich the case study of coefficient calibration for the AP in China and can improve the accuracy of calculating R_s and crop evapotranspiration based on existing data.     

FINDING CAUSES OF OUTLIERS IN MULTIVARIATE ENVIRONMENTAL DATA

Multivariate outliers in environmental data sets are often caused by atypical measurement error in a singlevariable.From a quality assurance perspective it is important to identify these variables efficiently so thatcorrective actions may be performed.We demonstrate a procedure for using two multivariate tests toidentify which variable‘caused’each outlier.The procedure is tested with simulated data sets that havethe same correlation structure as selected water chemistry variables from a survey of lakes in the WesternUnited States.The success rates are evaluated for three of the variables for sample sizes of 50 and 100,significance levels of 0.01 and 0.05 and various amounts of mean shift.The procedure works best forhighly correlated variables.     

REAL-TIME FILTERING OF DATA FROM MOBILE,PASSIVE REMOTE INFRARED SENSORS WITH PRINCIPAL COMPONENT MODELS OF BACKGROUND

Real-time monitoring of pollutant levels from a mobile measuring platform requires fast,flexible dataanalysis methods.This paper reports a method for rapid analysis of passive remotely sensed infrared datawith the aid of a Kalman filter.The background spectra produced by emission from the atmosphere aremodelled at the start of the data collection sequence with a simple principal components model obtainedby eigenanalysis of the initial‘blank’data taken with the spectrometer.The species of interest areincluded in the state space model by a separate measurement of their infrared spectra.It is demonstratedthat for best filter performance in detecting the simulated pollutant species SF_6 in the atmosphere,a filtermodel with two principal components describing the emission background works best.The filter‘maps’of SF_6 closely follow the integrated spectral intensities measured after removal of suitable backgrounds.     

SHORT COMMUNICATION USE OF METEOROLOGICAL DATA TO IDENTIFY EMISSOR SOURCES OF POLLUTANTS

A multivariate statistical analysis was performed by combining metal and anion data found in particulatematter from the atmosphere of Sant Adrià de Besòs(Barcelona)with SO_2,total suspended particle andsmoke values as well as with some meteorological measurements.Factor analysis allowed us in a previous work to recognize some emissor sources(soil,traffic,NO_x-SO_2,fuel oil,sea,incineration)by means of the analysis of some metals and anions.In the presentwork some meteorological data(temperature,humidity and sunshine),the concentration of SO_2 andvalues of total suspended particles and smoke were introduced into the correlation matrix in order tostudy their contribution to the identification of emissor sources.It was found that a decrease in temperature is in accordance with an increase in the presence of tracersfrom traffic emissions;chloride and humidity have a close correlation:SO_2 and sulphates come from thesame source;and oxidation from NO_x and SO_2 to nitrate and sulphate is higher with a decrease in thepercentage of sunshine.     

构造沉降和泥沙淤积对洞庭湖区防洪的影响

With the long-term data of the geodetic sea level measurements undertaken in the Dongting Basin and the recent sediment data of Dongting Lake, we analyze the tectonic subsidence rate of the Dongting Basin and the sedimentary rate of Dongting Lake. From the point of view of geomorphology and hydrogeology, we distinguish the two different spatial concepts between “the basin of Dongting Lake“ and “the Dongting Basin“. Then, we discuss the influences of the tectonic subsidence and the siltation on the levees and the space of storing flood. The better quality of levees is required due to the tectonic subsidence and the siltation, and the difficulties of preventing flood disasters are increasing.The space of storing flood is not affected by the tectonic subsidence, but by the siltation. At present,the sedimentary rate of Dongting Lake is higher than the tectonic subsidence rate of the Dongting Basin. The tectonic subsidence capacity of the Dongting Basin counteracts a part of sedimentary capacity, and the shrinking tendency of Dongting Lake is restrained to a certain extent, but the tectonic subsidence is harmful to the situation of preventing flood disasters in the Dongting Lake area.     

WHICH PRINCIPAL COMPONENTS TO UTILIZE FOR PRINCIPAL COMPONENT REGRESSION

Principal components(PCs)for principal component regression(PCR)have historically been selectedfrom the top down for a reliable predictive model.That is,the PCs are arranged in a list starting withthe most informative(PC associated with the largest singular value)and proceeding to the leastinformative(PC associated with the smallest singular value).PCs are then chosen starting at the top ofthis list.This paper discusses an alternative procedure of treating PC selection as an optimization prob-lem.Specifically,without any regard to the ordering,the optimal subset of PCs for an acceptablepredictive model is desired.Five data sets are analyzed using the conventional and alternative approaches.Two data sets are spectroscopic in nature,two data sets deal with quantitative structure-activityrelationships(QSARs)and one data set is concerned with modeling.All five data sets confirm thatselection of a subset without consideration to order secures the best results with PCR.One data set isalso compared using partial least squares 1.     

COASTAL PHENOMENA AND ISOSTATIC UPLIFT AROUND FILDES PENINSULA OF KING GEORGE ISLAND, SOUTH SHETLAND ISLANDS, ANTARCTICA

After a field work on recently raised coastal features in South Shetland Islands, we conclude that there are three kinds of recent coastal features, i. e, fragmental coast, rock coast, and ice cliff coast. The processes and forms of the fragmental cost formed by wave action with floating ice are studied. Systermtic study of the raised coasts enables us to distinguish two groups of the coasts by an elevation of 20m a. s. l. The raised coasts above 20m a. s. l. belong to older group and those below 20m a. s. l. belong to younger group. Mean uplift rate of coast in this area is 6.7 mm/a.     

A TEST OF SIGNIFICANCE FOR PARTIAL LEAST SQUARES REGRESSION

Partial least squares (PLS) regression is a commonly used statistical technique for performingmultivariate calibration, especially in situations where there are more variables than samples. Choosingthe number of factors to include in a model is a decision that all users of PLS must make, but iscomplicated by the large number of empirical tests available. In most instances predictive ability is themost desired property of a PLS model and so interest has centred on making this choice based on aninternal validation process. A popular approach is the calculation of a cross-validated r~2 to gauge howmuch variance in the dependent variable can be explained from leave-one-out predictions. Using MonteCarlo simulations for different sizes of data set, the influence of chance effects on the cross-validationprocess is investigated. The results are presented as tables of critical values which are compared againstthe values of cross-validated r~2 obtained from the user's own data set. This gives a formal test forpredictive ability of a PLS model with a given number of dimensions.     

Human’s digital footprints on the Qinghai-Tibet Plateau: Variations during festivals and impacts on nature reserves

Mobile internet and wireless communication technologies have produced unprecedented location-aware data. Such big geospatial data can be used as a proxy measure of the ‘digital footprints' left by us on the planet and provide a valuable opportunity to understand the dynamic and short-term human disturbance on the nature at fine scales. This study investigated the spatiotemporal variations of human's digital footprints on the Qinghai-Tibet Plateau using smartphone-users-generated Tencent's location request data. The results showed that human's digital footprints cover less than 5% of Qinghai and Tibet, exhibiting either a U-shaped or an N-shaped temporal change pattern during the major festivals. Spatial changes of the digital footprints manifested a transition process from dispersion to concentration in Xining and Lhasa. Human disturbance assessment of seven large nature reserves on the plateau showed that the Qinghai Lake is the most disturbed one as shown by 14.6% of its area is stained with human digital footprints and the areal average of footprint intensity is 1.59, and the disturbance was significantly escalated during the National Day holiday. By contrast, the Qangtang and Hoh Xil are the least affected nature reserves with the two indices less than 1% and 0.1, respectively.